CN1704990B - Information embedding device, information detecting device, information embedding and detecting system, information embedding method, information detecting method, information embedding program, infor - Google Patents
Information embedding device, information detecting device, information embedding and detecting system, information embedding method, information detecting method, information embedding program, infor Download PDFInfo
- Publication number
- CN1704990B CN1704990B CN2005100720253A CN200510072025A CN1704990B CN 1704990 B CN1704990 B CN 1704990B CN 2005100720253 A CN2005100720253 A CN 2005100720253A CN 200510072025 A CN200510072025 A CN 200510072025A CN 1704990 B CN1704990 B CN 1704990B
- Authority
- CN
- China
- Prior art keywords
- information
- value
- character
- characteristic quantity
- embedding
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K15/00—Arrangements for producing a permanent visual presentation of the output data, e.g. computer output printers
- G06K15/02—Arrangements for producing a permanent visual presentation of the output data, e.g. computer output printers using printers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/80—Recognising image objects characterised by unique random patterns
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Editing Of Facsimile Originals (AREA)
- Image Processing (AREA)
Abstract
An information embedding device is provided that embeds information in a document image having a character region therein. This information embedding device includes: a character region detecting unit that detects the character region from the document image; a character region dividing unit that divides the character region into a plurality of sub regions; and a feature quantity increasing/decreasing unit that performs an operation to increase or decrease a feature quantity with respect to each of the sub regions to a value that is determined from a pseudorandom number sequence in accordance with predetermined rules.
Description
Technical field
The present invention relates to a kind of information embedding device, the information detection device, information embeds and detection system, information embedding method, the information detection method, information embeds program, the information detection program, information embeds and locator(-ter), and recording medium.Especially, the present invention relates to a kind ofly, can not be identified by the human eye or be difficult to be identified by the human eye mode, in having the file and picture of character zone, embed the technology of information, and the invention still further relates to a kind of technology of surveying the information of this embedding with information.
Background technology
In recent years, can not have been developed the technology of the information that in such as the numerical data of view data and voice data, embeds by people's identification mode with information.By such technology, copyright information can embed in the data in advance, thereby can survey the owner of copyright from the copy of data.By this way, can prevent that the undelegated of numerical data that provides by the Internet from duplicating.Embed and the Detection Techniques field in the information that the present invention relates to, have following prior art.
Japanese Laid-Open Patent Application 2000-99501 has disclosed " embedding the method and system of information in document data ".In the prior art, disclosed the method for the information that in such as the document files of PDF or PS, embeds.Analyze the space of a whole page according to this invention, extract the character that is positioned at by a pseudo random number appointed positions, and the control relevant position.Then, carry out the embedding of information.
Japanese Laid-Open Patent Application 2001-78006 has disclosed " embedding and survey the method and apparatus of watermark information in the black and white binary document image ".In the prior art, disclosed and embedding and surveying for the information in the file and picture of bianry image.By the structure analysis of on file and picture, carrying out, image division is become a plurality of subdivisions, and two or more subdivisions form one group.By the characteristic quantity between more a plurality of groups, execution information embeds and surveys.
Japanese Laid-Open Patent Application 2002-232679 has disclosed " image processing method and equipment, computer program and recording medium ".In the prior art, by quantizing the ratio of adjacent character interval sum and adjacent character difference at interval, execution information embeds.
According to arbitrary foregoing invention,, extract each information embedded location accurately by execution architecture analysis on file and picture.Unnecessary embedded location is assigned to one group of information, thereby can carries out information detection very effectively.Yet, after the image that embeds information is printed, can scan and retrieve this information.Under these circumstances, the quantity that can embed information and the stability of information detection are provided with restriction.At this, there is the problem that the fiduciary level of detection information can not be measured equally.
Summary of the invention
Fundamental purpose of the present invention is to provide a kind of information that overcomes above-mentioned shortcoming to embed and Detection Techniques.
More particularly purpose of the present invention is, in the mode that information can not be identified by the human eye or be difficult to be identified by the human eye, provides a kind of technology that embeds information in having the file and picture of character zone.
Another special purpose of the present invention is to provide a kind of information embedding device, the information detection device, and information embeds and detection system, information embedding method, the information detection method, information embeds program, the information detection program, information embeds and locator(-ter), and recording medium.Use arbitrary said apparatus, method and program, information embed in the print image with being stabilized, and this information is accurately surveyed, and each user can know the reliability of the information of detection.
Above-mentioned purpose of the present invention realizes by a kind of information embedding device that embeds information in having the file and picture of character zone.This information embedding device comprises: a character zone probe unit, and it is from the document image detection character zone; One character zone division unit, it is divided into a plurality of subregions with this character zone; Increase and reduce the unit with a characteristic quantity, executable operations is to increase or to reduce the value of extremely determining from a pseudo-random number sequence according to predetermined rule about the characteristic quantity of each subregion.
In this information embedding device, this determines that the predetermined rule of the value of characteristic quantity comprises the multiple of suitable constant; And this information embedding device further comprises: a pseudo random number adder unit, it separates odd number multiple and even multiples according to the bit value that will embed information, and the value of pseudo random number is added to the value of this acquisition.
In this information embedding device, the character zone division unit is by selecting the external quadrilateral of paragraph, and one or its combination in the external quadrilateral on hurdle and the external quadrilateral of character are divided into a plurality of subregions with character zone.
In this information embedding device, the characteristic quantity of subregion is by the number of picture elements of particular color, the quantity of particular topology, line-spacing, adjacent character at interval, one or its combination expression in the character anglec of rotation and the external quadrilateral area.
Above-mentioned purpose of the present invention is embedded in the information detection device realization of the information in the file and picture equally by a kind of detection.This information detection device comprises: a character zone probe unit, and it is from the document image detection character zone; One character zone division unit, its character zone that will survey is divided into a plurality of subregions; One Characteristic Extraction unit, it extracts the characteristic quantity of subregion; With a value of information determining unit, it is by characteristic quantity that compares subregion and the value of determining embedding information from the value that a pseudo-random number sequence is determined.
In this information detection device, the character zone division unit is by selecting the external quadrilateral of paragraph, and one or its combination in the external quadrilateral on hurdle and the external quadrilateral of character are divided into a plurality of subregions with character zone.
Further comprise in this information detection device, a characteristic quantity correcting unit, it proofreaied and correct the characteristic quantity that is obtained by the Characteristic Extraction unit before value of information determining unit is determined.
In this information detection device, the characteristic quantity correcting unit comprises: a test pattern reading unit, and it reads the test pattern that is preset with the arbitrary characteristics amount; One correction function determining unit, it determines correction function based on the corresponding relation of the preset value in read value and the test pattern; With a correction operation unit, it uses correction function to carry out and proofreaies and correct.
This information detection device further comprises: a correction error probe unit, the error in its calculated correction function; With a mistake detectivity probe unit, it calculates for the mistake detectivity that embeds the information in the unknown images.
Above-mentioned purpose of the present invention embeds by a kind of information equally and detection system realizes, this system utilizes above-mentioned information embedding device to embed information, and utilizes the information detection device to carry out information detection.In this system, this information embedding device uses identical pseudo-random number sequence with this information detection device.
Above-mentioned purpose of the present invention realizes by a kind of information approach that embeds in having the file and picture of character zone equally.The method comprising the steps of: from the document image detection character zone; This character zone is divided into a plurality of subregions; With executable operations to increase or to reduce characteristic quantity about each subregion to the value of determining from a pseudo-random number sequence according to predetermined rule.
In the method, this determines that the predetermined rule of the value of characteristic quantity comprises the multiple of suitable constant.This method further comprises step: the value of pseudo random number is added to the value that odd number multiple and even multiples is separately obtained by according to the bit value that will embed information.
In the method, with the external quadrilateral of paragraph, one or its combination execution character area dividing step in the external quadrilateral on hurdle and the external quadrilateral of character.
In the method, the characteristic quantity of subregion is by the number of picture elements of particular color, the quantity of particular topology, line-spacing, adjacent character at interval, one or its combination expression in the character anglec of rotation and the external quadrilateral area.
Above-mentioned purpose of the present invention is embedded in the method realization of the information in the file and picture equally by a kind of detection.The method comprising the steps of: from the document image detection character zone; The character zone of this detection is divided into a plurality of subregions; Extract the characteristic quantity of subregion; With characteristic quantity by subregion relatively with determine the value of embedding information from the value that a pseudo-random number sequence is determined.
In the method, with the external quadrilateral of paragraph, one or its combination execution character area dividing step in the external quadrilateral on hurdle and the external quadrilateral of character.
This method further comprises step: proofread and correct the characteristic quantity that is obtained by the Characteristic Extraction step.This characteristic quantity aligning step was carried out before value of information determining step.
In the method, the characteristic quantity aligning step comprises step: read the test pattern that is preset with the arbitrary characteristics amount; Based on the corresponding relation of the preset value in read value and the test pattern, determine correction function; Carry out correction with the use correction function.
In the method, the characteristic quantity of subregion is by the number of picture elements of particular color, the quantity of particular topology, line-spacing, adjacent character at interval, one or its combination expression in the character anglec of rotation and the external quadrilateral area.
Above-mentioned purpose of the present invention is carried out by a kind of information embedding and detection method according to above-mentioned information embedding method and above-mentioned information detection method equally.At this, use identical pseudo-random number sequence in this information embedding method and this information detection method.
Above-mentioned purpose of the present invention realizes that by a kind of information embedding program this program makes computing machine carry out each step of above-mentioned information embedding method equally.
Above-mentioned purpose of the present invention realizes that by a kind of information detection program this program makes computing machine carry out each step of above-mentioned information detection method equally.
Above-mentioned purpose of the present invention embeds by a kind of information equally and locator(-ter) realizes, this program makes computing machine carry out each step of above-mentioned information embedding and detection method.
Above-mentioned purpose of the present invention realizes that by a kind of recording medium this recording medium can be by wherein can reading by stored program computing machine equally.
According to the present invention, information can be in its mode that can not be identified by the human eye or be difficult to be identified by the human eye, and embedding has in the file and picture of character zone.Equally, the information of this embedding can be detected.Information can stably embed one equally and print in the file and picture, and can be detected.In addition, the fiduciary level of detection information can be notified to the user.
More particularly, according to the present invention, thereby the structure of analytical documentation is only extracted character zone.This character zone is according to paragraph, and row and character are divided by known document analysis technology.Extract the unit area of the information that wherein will embed then.Usually, the image of printing because the position move or rotate and comprise distortion.Yet, can specify the information embedded location that is not subjected to deformation effect in the present invention.
After extracting the embedding zone, when controlling the characteristic quantity that calculates from the zone according to predetermined rule, embedding information.Because this rule need be difficult to for the 3rd people determine whether the information that embeds exists as the pseudo-random number sequence of parameter.
According to the present invention, according to embedding the identical rule of rule that extracted region is used in handling, the information of embedding can be detected.In the time will surveying the information of embedding, need to embed employed rule and pseudo-random number sequence.Thereby,, be difficult to survey the information of this embedding for the 3rd people.
Equally,, when information detection, use correction function according to the present invention, thus the characteristic quantity when information reproduction embeds, and this correction function is used to proofread and correct the image change that measures in advance based on reading device.This correction function is determined by reading the test pattern with known features amount in advance.By using the correct operation of correction function, can improve from the degree of accuracy of the image detection information of printing.
Further, according to the present invention, the error of the correction function that obtains when determining correction function is used for estimating the mistake detectivity.Thereby the user uses said apparatus and above-mentioned carrier, can know the fiduciary level of the information of surveying by said method.
Above and other objects of the present invention, feature and advantage can be from the detailed descriptions of the accompanying drawing of combination and are more obvious.
Description of drawings
Fig. 1 is the process flow diagram according to the operation of information embedding method of the present invention;
Fig. 2 is the process flow diagram according to information detection operation of the present invention;
Fig. 3 is a process flow diagram of determining operation according to correction function of the present invention;
Fig. 4 has shown the profile length in 3 * 3 piece;
Fig. 5 A has shown that information embeds image before;
Fig. 5 B has shown by embed information image afterwards according to information embedding method of the present invention;
Fig. 6 A has shown that information embeds image before;
Fig. 6 B has shown the image of the information that embeds and by according to the embedding operation of information embedding method of the present invention;
Fig. 7 has shown the information detection operation;
Fig. 8 has shown the definite operation of correction function;
Fig. 9 has shown mistake detectivity estimation operation; With
Figure 10 has exemplarily shown according to information embedding device of the present invention.
Embodiment
It below is description with reference to the embodiments of the invention of appended accompanying drawing.
Among the embodiment that is described below, described a kind of method that information is embedded in the black and white binary document image that comprises character and figure, this image only comprises the japanese sentence that horizontal bar is not set, and has described the device according to this method operation.In the following embodiments, being used for the characteristic quantity that information embeds is the quantity of every capable black picture element.
＜information embedding/detection method 〉
At first, describe according to information embedding method of the present invention and information detection method.
＜information embeds 〉
Fig. 1 is the process flow diagram according to information embedding operation of the present invention.This information embedding operation is carried out as follows.
In step S101, send the indication of document printing image.
In step S102, isolate character zone and graphics field from file and picture, character zone from this image detection to.
In step S103, character zone is divided into a plurality of subregions.In this embodiment, this is divided by row and carries out.Produce a histogram, this histogram has the black picture element that adds up in the horizontal direction.When this histogram surpasses a threshold value, be defined as existing histogram to surpass the row of this threshold value, the external quadrilateral that centers on this row is counted as a subregion.
In step S104, extract the characteristic quantity of subregion.Here, characteristic quantity is by the quantitaes of the black picture element in this zone, thereby black picture element is counted.
In step S105, generate a pseudo random number.The type stores of the sequence of this pseudo random number is as key information.
In step S106, the preset value of the value addition calculation characteristic quantity by the information that will embed and pseudo random number.More particularly, the pseudo random number that generates in step S105 is divided by " a ", and the remainder of Huo Deing is represented by " b " here.The value of characteristic quantity is " x ", and quantifying unit is represented by " q ".When information " 1 " during for the information that will embed, in the value that obtains by the product that adds " q " and odd number by " b " and " x " immediate value be set to preset value.When information " 0 " during for the information that will embed, in the value that obtains by the product that adds " q " and even number by " b " and " x " immediate value be set to preset value.
In step S107, characteristic quantity increased or be reduced to the preset value that calculates at step S106.Increase or reduce the quantity of black picture element, i.e. characteristic quantity, thus can reduce deterioration as far as possible to human eye.Here, the priority of pixel correction is determined by two following standards.1) priority should be had the piece of a plurality of pixels, this pixel is to change for the unconspicuous complex way of human eye.2) make correction with change as far as possible little on the complicacy.More particularly, this complicacy is represented by the profile length of four adjacent domains in 3 * 3 the piece.Profile length in this piece is set as L, and poor (absolute value) of the profile amount that changes along with the pixel inversion at piece center is Δ, and value J=L-6.5 * Δ is the priority of correction that the pixel of piece center is made.Fig. 4 has shown the example with L and Δ situation.
In Fig. 4, the part of being represented by circle (zero) is equivalent to profile, the quantitaes profile length of circle.At the figure of right-hand side is that the center pixel of figure by the counter-rotating left-hand side obtains.Under these circumstances, the difference Δ is 2, and priority can be calculated as 7-6.5 * 2=-6.
With above-described method calculating priority level, and repeat to revise, until the number of targets that obtains black picture element from BOB(beginning of block) with highest priority value.At this, under the situation that the number of black picture element will increase, only revise piece, and under the situation that the number of black picture element will reduce, only revise piece with black center with white center.
In step S108, determine whether in all subregions, to finish characteristic quantity control.
If finished characteristic quantity control in all subregions, operation moves to step S109.If the subregion that exists one or more characteristic quantity controls also not finish, step S104 is returned in operation.
In step S109, the file and picture of characteristic quantity control that printed executed, and EO.
Fig. 5 A and 5B have shown the example that embeds the file and picture of information front and back in top mode.
Fig. 6 A has shown the file and picture before the embedding information, and Fig. 6 B has shown an example, and at this, by this information embedding method, with quantifying unit q " 1000 ", pseudo-random number sequence bn " 20 ,-24 ,-3 " embeds Bit String " 111 ".
＜information detection 〉
Fig. 2 is the process flow diagram according to information detection method of the present invention.This information detection method comprises following step.
In step S201, test pattern is scanned instrument scanning, and reads in as file and picture.In this, use given threshold value to carry out binaryzation.
In step S202, go out character zone and graphics field from the document separation of images.
In step S203, the mode with identical with step S103 is divided into a plurality of subregions with character zone.
In step S204,, extract the characteristic quantity of subregion in the mode identical with step S104.
In step S205, use predetermined correction function, the value of correction feature amount.This correction function determines that method will describe at the step S301 to S309 of back.
In step S206, use the key of in step S105, storing to produce a pseudo random number.
In step S207, determine the value of embedding information.To be worth " x '-b " divided by the quantifying unit q identical with step S107, thereby obtain merchant " p " and remainder " r ", this value " x '-b " is to deduct the pseudo random number " b " that produces at step S206 and acquisition from the characteristic quantity of proofreading and correct among step S205 " x ' "." p " be odd number and " r " less than the situation of q/2 under or " p " for even number and " r " situation greater than q/2 under, determine the information of having embedded " 1 ".Not being under the above-mentioned situation, determine the information of having embedded " 0 ".In step S208, determine whether in all subregions, to have finished information and determine.Determine that if in all subregions, finished operation moves to step S209.If there are one or more definite subregions of determining also not finish, then operation is back to step S204.At step S209, the information that output is surveyed, thus this information can show on display device that or storage is hereof, or print.
Fig. 7 has shown after printing that scanning embeds the image of the information that Fig. 6 B is arranged and from an example of the operation of this information of image detection of scanning.In Fig. 7, use with embed processing in identical quantifying unit q=1000 and pseudo random number bn=" 20 ,-24 ,-3 ", detection information " 111 ".
＜determine the correction function of characteristic quantity 〉
Next, the correction function of describing step S205 is determined to handle.When scanning with reading images, each execution calculates this correction function.This correction function can forever use, unless very large change takes place the characteristic of reading device in the information detection step below once determining.Correction function determines that the process flow diagram of operation shows in Fig. 3, and the step of operation is described below.
In step S301, test pattern is scanned instrument scanning, and reads in as file and picture.In this, use and step S201 in identical threshold value, the execution binaryzation.This test pattern is the file and picture of black and white binaryzation, and the document image is with the form character display of japanese sentence and do not have level to write the hurdle.The document image is designed to have the wide distribution of the characteristic quantity (quantity of black picture element) on every hurdle.Equally, when producing, set in advance the value of characteristic quantity.
In step S302, go out character zone and graphics field from the document separation of images.
In step S303, in the mode identical with step S103, character zone is divided into a plurality of subregions.In step S304,, extract the characteristic quantity of subregion in the mode identical with step S104.
In step S305, determine whether in all subregions, to finish features extraction.If finished extraction in all subregions, this operation moves to step S306.If also have one or more subregions not finish extraction, then step S304 is returned in operation.
In step S306, by least squares approach, the mapping relations between the characteristic quantity of characteristic quantity that linear-apporximation is extracted in step S304 and the respective sub-areas when test pattern produces.Fig. 8 has shown an example of linear-apporximation.Be set to the correction function that uses in step S205 at the function of this acquisition.
In step S307, the quantifying unit " q " of using the sum of errors in the correction function that step S306 obtains to use in step S107 is calculated the mistake detectivity.Like method of estimation, calculate the probability density distribution of error in correction function by maximal phase, thereby the output valve of acquisition correction function exceeds the probability of this part ± q/2.This probability is set to the estimated value (referring to Fig. 9) of mistake detectivity.The value of this mistake detectivity is presented on the display device, thereby notifies the user.In addition, the value of this mistake detectivity is stored in the medium of scanner or information detection device, thereby the user can consult it at any time.
In step S308, the correction function that obtains at step S306 is stored in the medium of scanner or information detection device, thereby the information detection device can be consulted it in information detection.
＜information embedding/sniffer 〉
Figure 10 has described an example of the device that embodies above-described information embedding method and information detection method.
As shown in figure 10, personal computer is used as a multi-purpose computer 101.CPU102 controls each module in this multi-purpose computer 101, and can the program of executive logging on recording medium.Hard disk 103 memory image files, and record embodies the program of above-described information embedding method and information detection method.Removable recording medium 104 can be floppy disk for example, MO, CD-ROM/R-RW or DVD-ROM/R/RW.Similar with hard disk 103, these removable recording medium 104 memory image files, and recorded information embeds and locator(-ter).
Temporary program and the image of handling by CPU of RAM105.Driver 106 can be carried out the device that describes below and the exchanges data between the multi-purpose computer 101.Scanner 107 reads the image of the information that will embed, image that information has embedded and the test pattern that is used for determining correction function.Printer 108 is printed the image of the information that embedded.Monitor 109 shows the information or the mistake detectivity that for example will embed.Keyboard/mouse 110 is used for the execution of indication information embedding/locator(-ter), and imports the content of the information that will embed.
Utilize said structure, execution information in the following manner embeds.At first, utilize scanner 107 to read in image; Or playback record on hard disk 103 or removable recording medium 104 view data and be loaded among the RAM105.Activate application program by keyboard/mouse 110.In application program, select the information that to embed then, or pass through keyboard input information.Send the indication of embedding information then by keyboard/mouse 110, and embed information according to the program that embodies above-described information embedding method.The image recording that has embedded information or prints by printer 108 on hard disk 103 or removable recording medium 104.
Next, the information detection operation of carrying out by such structure is described.Utilize scanner 107 to read in the image of the information that will detect from it; Perhaps, playback record on hard disk 103 or removable recording medium 104 view data and be loaded among the RAM105.Activate application program by keyboard/mouse 110.When having sent the indication of embedding information, the program that embodies above-described information detection method is activated.The information that detects may be displayed on the monitor 109, is stored on hard disk 103 or the removable recording medium 104, or prints by printer 108.
Next, describe the characteristic quantity correction function of carrying out by such structure and determine operation.At first, activate application program by keyboard/mouse 110, utilize scanner 107 to read in test pattern, this test pattern is generated predetermined characteristic quantity.This view data that reads is loaded on RAM105, and the type or the ID that read the scanner of this test pattern are read into.When sending the indication of the definite operation of an execution correction function, embody above-described correction function and determine that the program of operation is activated, definite then correction function.The correction function that should determine is with to be used for the type or the ID of scanner of read test image relevant.Then, this correction function of determining is stored on hard disk 103 or the removable recording medium 104.If in the stage of back, use identical scanner to come detection information, then will consult this correction function.Equally, the definite error based on time strategic point when correction function is determined obtains calculates the mistake detectivity in this system.The value of this mistake detectivity may be displayed on the monitor 109, maybe can be stored in hard disk 103 or removable recording medium 104 that the type or the ID of scanner are associated on, maybe can print by printer 108.
According to the present invention, the manufacturing that information embedding device and information detection device can be separated from each other also can manufacture an integrated system.
Should be noted that: the present invention is not limited in the specific implementations that discloses above, without departing from the present invention, can carry out other change and modification.
The present invention is based on Japan of proposing on May 31st, 2004 formerly patented claim 2004-161091 number, the full content of this application at this as a reference.
Claims (13)
1. information embedding device, it is embedding information in having the file and picture of character zone, comprising:
The character zone probe unit is from the document image detection character zone;
The character zone division unit is divided into a plurality of subregions with this character zone;
Characteristic quantity increases and reduces the unit, and executable operations is to increase or to reduce the value of extremely determining from a pseudo-random number sequence according to predetermined rule about the characteristic quantity of each subregion, and wherein, this determines that the predetermined rule of the value of characteristic quantity comprises the multiple of suitable constant; With
The pseudo random number adder unit, it separates odd number multiple and even multiples according to the bit value that will embed information, and the value of pseudo random number is added to the value of this acquisition.
2. information embedding device as claimed in claim 1, wherein,
The character zone division unit is by selecting the external quadrilateral of paragraph, and one or its combination in the external quadrilateral on hurdle and the external quadrilateral of character are divided into a plurality of subregions with character zone.
3. information embedding device as claimed in claim 1, wherein,
The characteristic quantity of subregion is by the number of picture elements of particular color, the quantity of particular topology, line-spacing, adjacent character at interval, one or its combination expression in the character anglec of rotation and the external quadrilateral area.
4. information detection device, its detection is embedded in the information in the file and picture, comprising:
The character zone probe unit is from the document image detection character zone;
The character zone division unit is divided into a plurality of subregions with the character zone of this detection;
The Characteristic Extraction unit, it extracts the characteristic quantity of subregion;
Value of information determining unit, it is by characteristic quantity that compares subregion and the value of determining the information of embedding from the value that a pseudo-random number sequence is determined;
The characteristic quantity correcting unit before the determining of value of information determining unit, proofread and correct the characteristic quantity that is obtained by the Characteristic Extraction unit, and this characteristic quantity correcting unit comprises: the test pattern reading unit, read the test pattern that is preset with the arbitrary characteristics amount; The correction function determining unit based on the corresponding relation of the preset value in read value and the test pattern, is determined correction function; And correction operation unit, use correction function to carry out and proofread and correct;
The correction error probe unit, the error in the calculated correction function; With
Mistake detectivity probe unit calculates for the mistake detectivity that is embedded in the information in the unknown images.
5. information detection device as claimed in claim 4, wherein,
The character zone division unit is by selecting the external quadrilateral of paragraph, and one or its combination in the external quadrilateral on hurdle and the external quadrilateral of character are divided into a plurality of subregions with character zone.
6. an information embeds and detection system, and utilize information embedding device embedding information as claimed in claim 1 and utilize information detection device as claimed in claim 4 to carry out information detection,
This information embedding device uses identical pseudo-random number sequence with this information detection device.
7. method that embeds information in having the file and picture of character zone comprises step:
From the document image detection character zone;
This character zone is divided into a plurality of subregions; Executable operations is to increase or to reduce the value of extremely determining from a pseudo-random number sequence according to predetermined rule about the characteristic quantity of each subregion, and wherein, this determines that the predetermined rule of the value of characteristic quantity comprises the multiple of suitable constant; With
The value of pseudo random number is added to the value that odd number multiple and even multiples is separately obtained by according to the bit value that will embed information.
8. method as claimed in claim 7, wherein,
With the external quadrilateral of paragraph, one or its combination execution character area dividing step in the external quadrilateral on hurdle and the external quadrilateral of character.
9. method as claimed in claim 7, wherein,
The characteristic quantity of subregion is by the number of picture elements of particular color, the quantity of particular topology, line-spacing, adjacent character at interval, one or its combination expression in the character anglec of rotation and the external quadrilateral area.
10. a detection is embedded in the method for the information in the file and picture, comprises step:
From the document image detection character zone;
The character zone of this detection is divided into a plurality of subregions;
Extract the characteristic quantity of subregion;
By characteristic quantity that compares subregion and the value of determining the information of embedding from the value that a pseudo-random number sequence is determined; With
The characteristic quantity that correction obtains in the Characteristic Extraction step,
The characteristic quantity aligning step was carried out before value of information determining step,
Wherein, the characteristic quantity aligning step comprises step:
Read the test pattern that is preset with the arbitrary characteristics amount;
Based on the corresponding relation of the preset value in read value and the test pattern, determine correction function; With
Using correction function to carry out proofreaies and correct.
11. method as claimed in claim 10, wherein,
With the external quadrilateral of paragraph, one or its combination execution character area dividing step in the external quadrilateral on hurdle and the external quadrilateral of character.
12. method as claimed in claim 10, wherein:
The characteristic quantity of subregion is by the number of picture elements of particular color, the quantity of particular topology, line-spacing, adjacent character at interval, one or its combination expression in the character anglec of rotation and the external quadrilateral area.
13. one kind embeds and detection method according to the information of information embedding method as claimed in claim 7 and information detection method as claimed in claim 10, it has the identical pseudo-random number sequence of using in this information embedding method and this information detection method.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2004-161091 | 2004-05-31 | ||
| JP2004161091A JP4298588B2 (en) | 2004-05-31 | 2004-05-31 | Information detection apparatus and information detection method |
| JP2004161091 | 2004-05-31 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1704990A CN1704990A (en) | 2005-12-07 |
| CN1704990B true CN1704990B (en) | 2011-08-03 |
Family
ID=35424847
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2005100720253A Expired - Fee Related CN1704990B (en) | 2004-05-31 | 2005-05-26 | Information embedding device, information detecting device, information embedding and detecting system, information embedding method, information detecting method, information embedding program, infor |
Country Status (3)
| Country | Link |
|---|---|
| US (1) | US7627136B2 (en) |
| JP (1) | JP4298588B2 (en) |
| CN (1) | CN1704990B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007299321A (en) * | 2006-05-02 | 2007-11-15 | Ricoh Co Ltd | Information processor, information processing method, information processing program and information storage medium |
| JP4740166B2 (en) | 2007-01-31 | 2011-08-03 | 株式会社リコー | Image forming apparatus, image forming system, image forming method, program, and recording medium |
| JP4539756B2 (en) * | 2008-04-14 | 2010-09-08 | 富士ゼロックス株式会社 | Image processing apparatus and image processing program |
| JP5157774B2 (en) | 2008-09-17 | 2013-03-06 | 株式会社リコー | Image forming apparatus, image forming method, and program |
| JP2011201096A (en) | 2010-03-25 | 2011-10-13 | Brother Industries Ltd | Image recording apparatus |
| CN102567938B (en) * | 2010-12-23 | 2014-05-14 | 北大方正集团有限公司 | Watermark image blocking method and device for western language watermark processing |
| JP7180327B2 (en) | 2018-11-30 | 2022-11-30 | 株式会社リコー | System, image forming apparatus, method and program |
| JP7310252B2 (en) | 2019-04-19 | 2023-07-19 | 株式会社リコー | MOVIE GENERATOR, MOVIE GENERATION METHOD, PROGRAM, STORAGE MEDIUM |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1292908A (en) * | 1999-01-15 | 2001-04-25 | 皇家菲利浦电子有限公司 | Method and equipment for embedding watermarkds in images |
| CN1426223A (en) * | 2002-12-30 | 2003-06-25 | 上海交通大学 | Fragile digital water print embedding and extracting method based on address code |
| US20030158876A1 (en) * | 2002-02-21 | 2003-08-21 | Koninklijke Philips Electronics N.V. | On-line randomness test through overlapping word counts |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2687781B2 (en) | 1991-10-16 | 1997-12-08 | 凸版印刷株式会社 | Coordinate input device |
| JP3339918B2 (en) | 1993-06-10 | 2002-10-28 | 株式会社リコー | Image forming storage device |
| JP2000099501A (en) * | 1998-09-17 | 2000-04-07 | Internatl Business Mach Corp <Ibm> | Method and system for padding information into document data |
| JP3373811B2 (en) * | 1999-08-06 | 2003-02-04 | インターナショナル・ビジネス・マシーンズ・コーポレーション | Method and apparatus for embedding and detecting watermark information in black and white binary document image |
| JP2001186326A (en) | 1999-12-27 | 2001-07-06 | Ricoh Co Ltd | Method for embedding picture forgery verifying data, method and device for verifying forgery of picture and recording medium |
| JP2002232698A (en) | 2000-11-30 | 2002-08-16 | Kowa Co | Digital watermark embedding method and extraction method, and their devices |
| JP3728209B2 (en) | 2001-01-30 | 2005-12-21 | キヤノン株式会社 | Image processing method and apparatus, computer program, and storage medium |
| WO2003009580A1 (en) * | 2001-07-10 | 2003-01-30 | Kowa Co., Ltd. | Electronic watermark embedding method and extraction method and apparatus for them |
| JP4086103B2 (en) | 2002-01-25 | 2008-05-14 | 財団法人大阪産業振興機構 | Image data creation method, falsification detection method, image data creation device, falsification detection device, computer program, and recording medium |
| JP2005528649A (en) | 2002-06-03 | 2005-09-22 | コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ | Re-embedding digital watermarks in multimedia signals |
-
2004
- 2004-05-31 JP JP2004161091A patent/JP4298588B2/en not_active Expired - Fee Related
-
2005
- 2005-05-26 CN CN2005100720253A patent/CN1704990B/en not_active Expired - Fee Related
- 2005-05-27 US US11/138,520 patent/US7627136B2/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1292908A (en) * | 1999-01-15 | 2001-04-25 | 皇家菲利浦电子有限公司 | Method and equipment for embedding watermarkds in images |
| US20030158876A1 (en) * | 2002-02-21 | 2003-08-21 | Koninklijke Philips Electronics N.V. | On-line randomness test through overlapping word counts |
| CN1426223A (en) * | 2002-12-30 | 2003-06-25 | 上海交通大学 | Fragile digital water print embedding and extracting method based on address code |
Also Published As
| Publication number | Publication date |
|---|---|
| JP4298588B2 (en) | 2009-07-22 |
| US7627136B2 (en) | 2009-12-01 |
| CN1704990A (en) | 2005-12-07 |
| JP2005341493A (en) | 2005-12-08 |
| US20050264851A1 (en) | 2005-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4926004B2 (en) | Document processing apparatus, document processing method, and document processing program | |
| JP5736767B2 (en) | Providing method of digital media, print medium, and computer program | |
| CN101160950B (en) | Image processing device, image processing method | |
| JP2007200014A (en) | Information processing device, information processing method, information processing program, and recording medium | |
| CN101155239A (en) | Electronic watermark embedding device and detection device | |
| CN103339920A (en) | Image processing apparatus and method of supplementing pixel value | |
| JP2004158036A (en) | Computer system for identifying area on instance of machine-readable form | |
| US20080174101A1 (en) | Counterfeit deterrence using dispersed miniature security marks | |
| CN1704990B (en) | Information embedding device, information detecting device, information embedding and detecting system, information embedding method, information detecting method, information embedding program, infor | |
| MX2008000767A (en) | System and method for embedding dispersed miniature security marks. | |
| JP4631900B2 (en) | Information processing apparatus, information processing system, and information processing program | |
| JPH08335266A (en) | Processor and method for image processing | |
| CN102171723A (en) | Method for red-eye detection | |
| CN100486296C (en) | Information extracting method, information extracting device, information extracting program and storage medium | |
| JP4440283B2 (en) | Image processing apparatus, control method thereof, control program, and storage medium | |
| US20030044084A1 (en) | Two dimensional image recording and reproducing scheme using similarity distribution | |
| JP2006252562A (en) | Image recognition method | |
| CN114298952A (en) | Label image generation method, device, equipment and storage medium | |
| JP2010102734A (en) | Image processor and program | |
| CN101025604A (en) | Management information adding method and image forming apparatus | |
| CN1326383C (en) | Image processing apparatus, image processing method, computer program and computer readable storage medium | |
| JP5128563B2 (en) | Document verification system, document verification method, document verification program, and recording medium | |
| JP4552757B2 (en) | Image processing apparatus, image processing method, and image processing program | |
| JP5298830B2 (en) | Image processing program, image processing apparatus, and image processing system | |
| JP2010103862A (en) | Document processing apparatus and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110803 Termination date: 20190526 |